Managing energy, performance and cost in large scale heterogeneous datacenters using migrations

Abstract Improving datacenter energy efficiency becomes increasingly important due to energy supply problems, fuel costs and global warming. Virtualisation can help to improve datacenter energy efficiency through server consolidation which involves migrations that can be expensive in terms of extra energy consumption and performance loss. This is because, in clouds, Virtual Machines (VMs) of the same instance class running on different hosts may perform quite differently due to resource heterogeneity. As a result of variations in performance, different runtimes will exist for a given workload, with longer runtimes potentially leading to higher energy consumption. For a large datacenter, this would both reduce the overall throughput, and increase overall energy consumption and costs. In this paper, we demonstrate how the performance of workloads across different CPU models leads to variability in energy efficiencies, and therefore costs. We investigate through a number of experiments, using the Google workload traces for 12,583 hosts and 492,309 tasks, the impact of migration decisions on energy efficiency when performance variations of workloads are taken into account. We discuss several findings, including (i) the existence of a trade-off between overall energy consumption and performance (hence cost), (ii) that higher utilization decreases the energy efficiency as it offers fewer chances to CPU management tools for energy savings, and (iii) how our migration approach could save up to 3.66% energy, and could improve VMs performance up to 1.87% compared with no migration. Similarly, compared with migrate all, the proposed migration approach could save up to 2.69% energy, and improve VMs performance up to 1.01%. We discuss these results for different combinations of VM allocation, migration policies and different benchmark workloads. 1

[1]  Chita R. Das,et al.  Managing performance and energy in large scale data centers , 2012 .

[2]  Albert Y. Zomaya,et al.  Profiling-Based Workload Consolidation and Migration in Virtualized Data Centers , 2015, IEEE Transactions on Parallel and Distributed Systems.

[3]  Albert G. Greenberg,et al.  The cost of a cloud: research problems in data center networks , 2008, CCRV.

[4]  Athanasios V. Vasilakos,et al.  Managing Performance Overhead of Virtual Machines in Cloud Computing: A Survey, State of the Art, and Future Directions , 2014, Proceedings of the IEEE.

[5]  Rajkumar Buyya,et al.  Energy-aware resource allocation heuristics for efficient management of data centers for Cloud computing , 2012, Future Gener. Comput. Syst..

[6]  Tommaso Cucinotta,et al.  The effects of scheduling, workload type and consolidation scenarios on virtual machine performance and their prediction through optimized artificial neural networks , 2011, J. Syst. Softw..

[7]  Rajkumar Buyya,et al.  Cost of Virtual Machine Live Migration in Clouds: A Performance Evaluation , 2009, CloudCom.

[8]  Radu Prodan,et al.  Modelling energy consumption of network transfers and virtual machine migration , 2016, Future Gener. Comput. Syst..

[9]  Lee Gillam,et al.  Performance Evaluation for Cost-Efficient Public Infrastructure Cloud Use , 2014, GECON.

[10]  Akshat Verma,et al.  pMapper: Power and Migration Cost Aware Application Placement in Virtualized Systems , 2008, Middleware.

[11]  Hai Jin,et al.  Heterogeneity and Interference-Aware Virtual Machine Provisioning for Predictable Performance in the Cloud , 2016, IEEE Transactions on Computers.

[12]  Randy H. Katz,et al.  Above the Clouds: A Berkeley View of Cloud Computing , 2009 .

[13]  Gautam Kar,et al.  Application Performance Management in Virtualized Server Environments , 2006, 2006 IEEE/IFIP Network Operations and Management Symposium NOMS 2006.

[14]  Stephen W. Poole,et al.  Towards efficient supercomputing: searching for the right efficiency metric , 2012, ICPE '12.

[15]  Lee Gillam,et al.  Sibling virtual machine co-location confirmation and avoidance tactics for Public Infrastructure Clouds , 2016, The Journal of Supercomputing.

[16]  Arun Venkataramani,et al.  Sandpiper: Black-box and gray-box resource management for virtual machines , 2009, Comput. Networks.

[17]  Randy H. Katz,et al.  Heterogeneity and dynamicity of clouds at scale: Google trace analysis , 2012, SoCC '12.

[18]  Boon Thau Loo,et al.  Exploiting Cloud Heterogeneity to Optimize Performance and Cost of MapReduce Processing , 2015, PERV.

[19]  Ayse K. Coskun,et al.  Energy-efficient server consolidation for multi-threaded applications in the cloud , 2013, 2013 International Green Computing Conference Proceedings.

[20]  Radu Prodan,et al.  A Workload-Aware Energy Model for Virtual Machine Migration , 2015, 2015 IEEE International Conference on Cluster Computing.

[21]  César A. F. De Rose,et al.  Server consolidation with migration control for virtualized data centers , 2011, Future Gener. Comput. Syst..

[22]  Albert Y. Zomaya,et al.  A Taxonomy and Survey of Energy-Efficient Data Centers and Cloud Computing Systems , 2010, Adv. Comput..

[23]  Luiz Fernando Bittencourt,et al.  Power-aware virtual machine scheduling on clouds using active cooling control and DVFS , 2011, MGC '11.

[24]  Rajkumar Buyya,et al.  Energy Efficient Resource Management in Virtualized Cloud Data Centers , 2010, 2010 10th IEEE/ACM International Conference on Cluster, Cloud and Grid Computing.

[25]  Lee Gillam,et al.  Re-appraising instance seeking in Public Clouds , 2015, 2015 Science and Information Conference (SAI).

[26]  Ian Sommerville,et al.  Understanding Tradeoffs between Power Usage and Performance in a Virtualized Environment , 2013, 2013 IEEE Sixth International Conference on Cloud Computing.

[27]  Andrzej Kochut,et al.  Dynamic Placement of Virtual Machines for Managing SLA Violations , 2007, 2007 10th IFIP/IEEE International Symposium on Integrated Network Management.

[28]  Mor Harchol-Balter,et al.  Stochastic Models and Analysis for Resource Management in Server Farms , 2011 .

[29]  Bo Li,et al.  iAware: Making Live Migration of Virtual Machines Interference-Aware in the Cloud , 2014, IEEE Transactions on Computers.

[30]  Luiz André Barroso,et al.  The Case for Energy-Proportional Computing , 2007, Computer.

[31]  Hermann de Meer,et al.  Modelling and analysing the power consumption of idle servers , 2012, 2012 Sustainable Internet and ICT for Sustainability (SustainIT).

[32]  Karl Aberer,et al.  Impact of Instance Seeking Strategies on Resource Allocation in Cloud Data Centers , 2013, 2013 IEEE Sixth International Conference on Cloud Computing.

[33]  Rajkumar Buyya,et al.  Managing Overloaded Hosts for Dynamic Consolidation of Virtual Machines in Cloud Data Centers under Quality of Service Constraints , 2013, IEEE Transactions on Parallel and Distributed Systems.

[34]  Richard E. Brown,et al.  United States Data Center Energy Usage Report , 2016 .

[35]  Thomas F. Wenisch,et al.  PowerNap: eliminating server idle power , 2009, ASPLOS.

[36]  Jerome A. Rolia,et al.  Resource Contention Detection in Virtualized Environments , 2015, IEEE Transactions on Network and Service Management.

[37]  Mor Harchol-Balter,et al.  Energy-Efficient Dynamic Capacity Provisioning in Server Farms , 2010 .

[38]  Charles Reiss,et al.  Towards understanding heterogeneous clouds at scale : Google trace analysis , 2012 .

[39]  Rizos Sakellariou,et al.  A Cloud Controller for Performance-Based Pricing , 2015, 2015 IEEE 8th International Conference on Cloud Computing.

[40]  Quanyan Zhu,et al.  Dynamic energy-aware capacity provisioning for cloud computing environments , 2012, ICAC '12.

[41]  Lee Gillam,et al.  Towards Performance Prediction for Public Infrastructure Clouds: An EC2 Case Study , 2013, 2013 IEEE 5th International Conference on Cloud Computing Technology and Science.

[42]  Rajkumar Buyya,et al.  CloudSim: a toolkit for modeling and simulation of cloud computing environments and evaluation of resource provisioning algorithms , 2011, Softw. Pract. Exp..

[43]  Rajkumar Buyya,et al.  Adaptive threshold-based approach for energy-efficient consolidation of virtual machines in cloud data centers , 2010, MGC '10.

[44]  Calton Pu,et al.  Mistral: Dynamically Managing Power, Performance, and Adaptation Cost in Cloud Infrastructures , 2010, 2010 IEEE 30th International Conference on Distributed Computing Systems.

[45]  Lee Gillam,et al.  Should infrastructure clouds be priced entirely on performance? An EC2 case study , 2014, Int. J. Big Data Intell..

[46]  Michela Meo,et al.  Probabilistic Consolidation of Virtual Machines in Self-Organizing Cloud Data Centers , 2013, IEEE Transactions on Cloud Computing.

[47]  Mehul A. Shah,et al.  Analyzing the energy efficiency of a database server , 2010, SIGMOD Conference.

[48]  Michela Meo,et al.  Analysis of a Self-Organizing Algorithm for Energy Saving in Data Centers , 2013, 2013 IEEE International Symposium on Parallel & Distributed Processing, Workshops and Phd Forum.

[49]  Henry Hoffmann,et al.  Minimizing energy under performance constraints on embedded platforms: resource allocation heuristics for homogeneous and single-ISA heterogeneous multi-cores , 2015, SIGBED.

[50]  Karim Djemame,et al.  Energy Efficiency Support Through Intra-layer Cloud Stack Adaptation , 2016, GECON.

[51]  Lee Gillam,et al.  An Energy Aware Cost Recovery Approach for Virtual Machine Migration , 2016, GECON.

[52]  Chita R. Das,et al.  A dynamic energy management scheme for multi-tier data centers , 2011, (IEEE ISPASS) IEEE INTERNATIONAL SYMPOSIUM ON PERFORMANCE ANALYSIS OF SYSTEMS AND SOFTWARE.